An improved approach to tuning MPC-PID controller parameters for non-Gaussian systems

In this paper, a data-driven approach to tuning PID parameters is presented for non-Gaussian systems. Model predictive control (MPC) algorithm is used to formulate the PID parameters tuning problem. The survival information potential (SIP) is employed to constructed the performance index to optimize PID parameters for non-Gaussian systems. The proposed PID controller parameter tuning method is then applied to control the superheating of an organic Rankine cycle based power plant, simulation results show the proposed method may obtain satisfactory control performance.